MATAKZINFinancial Advisory Group

Insights § 5

Plain answers on AI rules and AI costs

Short briefings for owners, administrators and CFOs. Status as of October 2026; we review these every quarter.

HEALTHCARE · OCT 2026

Texas medical practices: the AI disclosure duty you may be missing

If you use an AI scribe or chatbot, Texas already expects you to tell patients.

Since January 1, 2026, the Texas Responsible Artificial Intelligence Governance Act (TRAIGA) has required healthcare providers that use AI in services or treatment to disclose it to the patient or their representative. The disclosure must be clear, conspicuous and in plain language, and given no later than the date the service is first provided. In an emergency, it is due as soon as reasonably possible.

The Texas Attorney General enforces the law; there is no private right of action. A separate Texas law, SB 1188, adds expectations that practitioners review AI used in diagnosis and treatment.

Three questions to ask this week:

  • Is AI disclosure language in your intake paperwork and patient portal?
  • Does every AI vendor that touches patient data have a signed business associate agreement?
  • Does a clinician review every AI-drafted note before it enters the record?

If any answer is "not sure", the AI Risk Snapshot will tell you in 10 business days. Confirm final disclosure wording with your healthcare counsel.

BANKING · OCT 2026

SR 26-2 replaced SR 11-7. What it means for community institutions.

The first rewrite of model risk guidance in 15 years, with AI guidance still to come.

On April 17, 2026, the Federal Reserve, OCC and FDIC issued SR 26-2, revised guidance on model risk management that replaces SR 11-7. It keeps the familiar discipline: a model inventory, validation, ongoing monitoring and oversight of vendor models, scaled to the institution's size and risk.

The guidance is aimed mainly at larger banks, but examiners and boards use it as the benchmark everywhere. The agencies have said separate guidance on generative and agentic AI will follow.

What a community bank or credit union can do now:

  • Put models and AI tools in one inventory, including AI features inside vendor software
  • Tier them by materiality: credit, pricing, BSA/AML and fraud first
  • Require vendor documentation on how models work and change
  • Report AI risk to the board every quarter
FINANCE · OCT 2026

What is AI actually costing your business?

AI spend without a budget is a finance problem. AI without an owner is a risk problem.

Ask an owner what the business spends on AI and the usual answer is "not much." The card statements tell a different story: overlapping chatbot subscriptions, AI add-ons switched on inside existing software, and free tools nobody approved that quietly turned into paid plans.

A useful AI budget does three things. It lists every AI cost in one place. It ties each tool to a use case with a measurable result, such as hours saved or faster billing. And it pairs every tool with its risk rating, so the cheapest tool is never the one leaking client data.

That is why MATAKZIN reviews AI spend and AI risk together: the same ledger that shows what you pay also shows which tools IT never knew about.

Insights are general information, not legal advice. Confirm how a rule applies to you with your counsel.

Start with a 30-minute conversation

We will talk through how AI shows up in your business and where finance and risk fit in. If a Snapshot makes sense, you will have a proposal the same day.